Self-adaptive radio frequency interference intensity determination method for microwave radiometer

Abstract

The invention discloses a self-adaptive radio frequency interference intensity determination method for a microwave radiometer, and belongs to the technical field of satellite microwave ocean remote sensing. The method comprises the following steps of: performing adaptive scene analysis and judgment according to a brightness temperature image subjected to RFI pollution snapshot: dividing a background region of an RFI source on the brightness temperature image into a flat scene background and a non-flat scene background; the distance between the two RFI sources and the ground pixel resolution are determined according to user requirements and the height of the orbit where the satellite is located, if the distance between the two RFI sources does not exceed the ground pixel resolution, the two RFI sources are determined to be a close-distance RFI source, and otherwise, the two RFI sources are determined to be a long-distance RFI source; and for the RFI sources of different combination types, different intensity determining algorithms are adopted to obtain the intensity of the RFI sources. According to the method, the universality and the stability of the RFI strength determination method can be improved.

Description

Technical Field

[0001] The present invention relates to a method for adaptively determining the intensity of radio frequency interference (RFI) in a microwave radiometer, belonging to the technical field of satellite microwave ocean remote sensing. Background Art

[0002] Regarding the problem of determining the intensity of radio frequency interference (abbreviated as RFI determination of intensity), the industry has explored and developed various methods for determining the intensity based on the rich remote sensing data of the SMOS satellite, such as the point target method for determining the intensity and the step-by-step RFI method for determining the intensity, etc. From the research and analysis at home and abroad, it can be seen that the methods for positioning, determining the intensity, and suppressing RFI mainly face three challenges: 1) Two or more closely spaced RFIs; 2) A complex RFI environment in the land-sea boundary area; 3) Two or more RFIs with uneven strength distribution. Although corresponding solutions have been proposed for these challenges in existing research, however, these methods lack universality and stability in practical applications.

[0003] Currently, there is no RFI determination of intensity method that can comprehensively and effectively address the 3 challenges simultaneously, resulting in the lack of universality of the existing methods for determining the intensity in practical applications; when facing complex scenarios that simultaneously contain at least two challenges, the existing methods often cannot take into account all RFI sources and can only effectively process some of the RFIs, resulting in the lack of stability of the existing methods in complex scenarios. Summary of the Invention

[0004] The technical problem solved by the present invention is: overcoming the deficiencies of the prior art, a method for adaptively determining the intensity of radio frequency interference in a microwave radiometer is proposed. By classifying RFI sources and designing different algorithms for determining the intensity for different types of RFI sources, the universality and stability of the RFI determination of intensity method can be improved.

[0005] The technical solution of the present invention is:

[0006] A method for adaptively determining the intensity of radio frequency interference in a microwave radiometer, comprising:

[0007] Performing adaptive scene analysis and judgment based on the brightness temperature image of the snapshot contaminated by RFI: classifying the background area where the RFI sources are located on the brightness temperature image into a flat scene background and a non-flat scene background; pairing all RFI sources in pairs, and determining the distance between any pair of RFI sources and the ground pixel resolution according to user requirements and the orbital altitude of the satellite. If the distance does not exceed the ground pixel resolution, it is determined as a closely spaced RFI source, otherwise it is a remotely spaced RFI source;

[0008] For RFI sources in a flat scene background, the method for obtaining the intensity of the RFI source includes:

[0009] For each pair of RFI sources, the following processing is performed: For a nearby RFI source, the step-by-step RFI intensity determination algorithm is used to obtain the RFI source intensity; for a distant RFI source, the independent point target intensity determination algorithm is used to obtain the RFI source intensity.

[0010] For an RFI source in a non-flat scene background, the methods for obtaining the RFI source intensity include:

[0011] For each pair of RFI sources, the following processing is performed: For a distant RFI source, the RFI intensity determination algorithm near the coastline is used to obtain the RFI source intensity; for a nearby RFI source, the decorrelation RFI intensity determination algorithm is used to obtain the RFI source intensity.

[0012] If multiple intensity data are obtained for the same RFI source, the exact intensity is obtained by taking the average.

[0013] Furthermore, the background areas where the RFI sources are located on the brightness temperature image are divided into flat scene backgrounds and non-flat scene backgrounds. The specific method is as follows:

[0014] According to the position (ξ i , η i ) of the i-th RFI source in the brightness temperature image, project (ξ i , η i ) onto the Earth coordinate system to obtain the longitude and latitude coordinates of the RFI source as O(lon i , lat i ); with O(lon i , lat i ) as the center, the area delineated by a circle with a set distance r as the radius is the buffer zone. If the number of intersection points between the buffer zone and the coastline γ curve is greater than or equal to 2, it is determined to be a coastal area, belonging to an RFI source on a non-flat scene background; if the number of intersection points between the buffer zone and the coastline γ curve is less than 2, it is determined to be an inland sea or inland area, belonging to an RFI source on a flat scene background.

[0015] Furthermore, according to the user's requirements and the orbital altitude of the satellite, the distance between two RFI sources is determined. The determination method is as follows:

[0016] Obtain the initial positioning coordinates of the RFI source in the antenna field-of-view coordinate system according to the brightness temperature threshold method; combine the orbital altitude and attitude information of the satellite to convert the initial positioning coordinates in the synthetic aperture radiation coordinate system to the coordinates in the Earth longitude and latitude coordinate system.

[0017] Calculate the distance d between the two RFI sources:

[0018] d = R * arccos[cos(lon RFIi ) * cos(lon RFIj ) * cos(latRFIi -lat RFIj )

[0019] +sin(lat RFIi )*sin(lat RFIj )]

[0020] where R is the radius of the Earth, (lon RFIi , lat RFIi ) is the longitude and latitude of the i-th RFI source, and (lon RFIj , lat RFIj ) is the longitude and latitude of the j-th RFI source.

[0021] Furthermore, two ground pixel resolutions Res are determined according to the user requirements and the orbital altitude of the satellite. The method for determining Res is as follows:

[0022]

[0023] where D xmax represents the maximum baseline of the synthetic aperture radiometer antenna array in the X direction in the Earth-fixed coordinate system, D ymax represents the maximum baseline distance of the synthetic aperture radiometer antenna array in the Y direction in the Earth-fixed coordinate system, and (x sat , y sat , z sat ) represents the orbital position information of the satellite in the Earth-fixed coordinate system.

[0024] Furthermore, for a distant RFI source in a flat scene background, an independent point target calibration strength algorithm is used to obtain the RFI source strength. The specific method is as follows:

[0025] 1) Perform brightness temperature inversion on the original visibility measured by the synthetic aperture radiometer to obtain a brightness temperature image, and find the positions and intensities of the brightness temperature peaks in the brightness temperature image;

[0026] 2) Judge each brightness temperature peak in turn: If the intensity of the brightness temperature peak does not exceed 350K, there is no RFI source in the scene, and step 4) is executed; otherwise, there is an RFI source at the position;

[0027] 3) For each existing RFI source, perform the following processing:

[0028] Calculate the 1K response visibility and 1K inversion brightness temperature of each RFI source to obtain the intensity of each RFI source;

[0029] Calculate the scene visibility after suppressing all RFI sources according to the intensity of each RFI source, and return to step 1);

[0030] 4) Generate a position determination list of RFI sources {(ξ RFI,1 , ηRFI,1 ),…,(ξ RFI,q ,η RFI,q )}, perform fixed-strength determination on each RFI source in the RFI list one by one:

[0031] Calculate the average scene brightness temperature after suppressing the current RFI source, and then calculate the ratio of the brightness temperature after suppressing the RFI source to the average scene brightness temperature; when the ratio is less than 1, it means that the influence of the current RFI source on the scene has been completely suppressed, and the intensity of the current RFI can be obtained; otherwise, continue to iteratively suppress the current RFI source until the ratio is less than 1, and then obtain the intensity of the current RFI.

[0032] Furthermore, for a short-distance RFI source in a flat scene background, use a step-by-step RFI fixed-strength algorithm to obtain the RFI source intensity. The specific method is: suppress the intensity contribution of the RFI at the estimated position from the observation data, so as to extract the position and intensity of the signal source. Repeat the position determination and intensity determination of the RFI in this way until the amplitude intensity of the RFI is repeatedly searched

[0033] Furthermore, for a long-distance RFI source in a non-flat scene background, use an RFI fixed-strength algorithm near the coastline to obtain the RFI fixed-strength affected by the Gibbs oscillation at the land-sea boundary and the RFI source. The specific method is:

[0034] Calculate the average scene brightness temperature after suppressing the RFI source according to the actual scene brightness temperature jointly composed of the scene background brightness temperature and the RFI source brightness temperature;

[0035] Minimize the following formula according to the average scene brightness temperature after suppressing the RFI source to obtain the final goal of the RFI fixed-strength method near the coastline and achieve fixed-strength:

[0036]

[0037] In the formula, N land 、N sea represent the numbers of RFI interferences on land and sea; σ land 、σ sea represent the standard deviations of land and sea, which are obtained from the average scene brightness temperature after suppressing the RFI source.

[0038] Furthermore, for a short-distance RFI source in a non-flat scene background, use a decoherence RFI fixed-strength algorithm. By comparing the difference between the original brightness temperature and the inverted brightness temperature, a system of equations with the RFI source intensity as the unknown is constructed, and the fixed-strength of the short-distance RFI source is achieved by solving the system of equations.

[0039] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above method are implemented.

[0040] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above method are implemented.

[0041] The advantages of the present invention compared with the prior art are as follows:

[0042] (1) Compared with the existing independent point target calibration strength method, step-by-step RFI calibration strength method, RFI calibration strength method near the coastline, de-coherence RFI calibration strength method, etc., the adaptive radio frequency interference calibration strength method proposed by the present invention performs adaptive RFI calibration in complex scenarios such as flat and non-flat, far and near distances for one-dimensional synthetic aperture radiometers and two-dimensional synthetic aperture radiometers, and shows the smallest calibration strength error, fully verifying its universality.

[0043] (2) For a complex scenario that simultaneously includes three situations: "two close RFI", "complex RFI environment in the land-sea boundary area", and "two RFI with different intensities", on the basis of maintaining the performance of other RFI sources not lower than that of other similar calibration strength methods, the present invention always shows better performance than other methods in the calibration strength error of at least one RFI source, and this characteristic significantly enhances the overall stability and reliability of the RFI calibration strength method. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] By reading the following detailed description of the preferred embodiments, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered to be a limitation of the present invention. Moreover, throughout the drawings, the same reference numerals are used to represent the same components. In the drawings:

[0045] Figure 1 is a flowchart of the adaptive radio frequency interference calibration strength method for the microwave radiometer of the present invention;

[0046] Figure 2 is a relationship diagram of the RFI point source buffer area and the coastline of the present invention;

[0047] Figure 3 is a flowchart of the simulation analysis and verification of the adaptive RFI calibration strength selection method of the present invention;

[0048] Figure 4 is the original brightness temperature map of the simulation RFI scenario 1 of the embodiment of the present invention;

[0049] Figure 5This is the original brightness temperature map of the simulated RFI scenario 2 in the embodiment of the present invention;

[0050] Figure 6 This is the original brightness temperature map of the simulated RFI scenario 3 in the embodiment of the present invention;

[0051] Figure 7 This is the original brightness temperature map of the simulated RFI scenario 4 in the embodiment of the present invention;

[0052] Figure 8 This is the original brightness temperature map of the simulated RFI scenario 5 in the embodiment of the present invention. Detailed implementation manners

[0053] Hereinafter, exemplary embodiments of the present disclosure will be described in more detail with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be fully conveyed to those skilled in the art.

[0054] In view of the fact that current various RFI intensity determination methods only focus on specific scenarios and generally lack broad applicability across scenarios, the present invention proposes an adaptive radio frequency interference intensity determination method for microwave radiometers, which is applicable to RFI intensity determination data processing carried out by an aperture synthesis radiometer.

[0055] This method is as Figure 1 shown and includes the following steps:

[0056] S1: Adaptive scenario analysis and judgment

[0057] 1) Analyze the distance between the RFI and the coastline, and classify the RFI source area into two major categories: flat scene background or non-flat scene background.

[0058] The coastline is the coastline determined based on the tidal model and DEM technology, and is geometrically represented by a γ curve. The position of the i-th RFI is (ξ i , η i ). Projecting (ξ i , η i ) onto the earth coordinate system gives the longitude and latitude coordinates of the RFI as O(lon i , lat i ).

[0059] For the non-flat scene background, with O(lon i , lat i) centered at a point, with a buffer zone of a certain distance r as the radius (r can be adjusted according to the spatial resolution of the orbiting satellite. For example, for the SMOS satellite, r can be defined as more than 20 km). If the number of intersection points between the buffer zone and the coastline γ curve is greater than or equal to 2, it is considered a coastal area (land-sea boundary area), which is a non-flat area. If the number of intersection points between the buffer zone and the coastline γ curve is less than 2, it is considered an inland (sea) area, which is a flat area. Taking Figure 2 For example, for the two RFI sources in

[0060] 2) Classify according to the distance between RFI sources in the scene, into close-range RFI sources and long-range RFI sources

[0061] The initial positioning coordinates (ξ RFIi , η RFIi ) of the RFI obtained by the brightness temperature threshold method in the field-of-view coordinate system of the synthetic aperture radiometer antenna. According to information such as the orbital altitude and attitude of the satellite, the (ξ, η) coordinates in the field-of-view coordinate system of the synthetic aperture radiometer antenna are converted to the Earth's longitude and latitude coordinate system (lon RFIi , lat RFIi ). According to the longitude and latitude coordinates of any two RFIs on the Earth, the distance d between the two RFIs can be calculated:

[0062] d = R * arccos[cos(lon RFIi ) * cos(lon RFIj ) * cos(lat RFIi - lat RFIj )

[0063] + sin(lat RFIi ) * sin(lat RFIj )]

[0064] Among them, R is the Earth's radius of 6371 km, lon RFIi is the longitude (abscissa) of the i-th RFI, and lat RFIi is the latitude (ordinate) of the i-th RFI.

[0065] The judgment basis for two adjacent RFIs on the Earth is that the distance d between the two RFIs < the resolution Res of 2 ground pixels of the synthetic aperture radiometer. The calculation formula for Res is:

[0066]

[0067] Among them, D xmaxDenote the maximum baseline of the synthetic aperture radiometer antenna array in the X direction in the earth-fixed coordinate system as D ymax Denote the maximum baseline distance of the synthetic aperture radiometer antenna array in the Y direction in the earth-fixed coordinate system as (x sat ,y sat ,z sat ), which represents the orbital position information of the satellite in the earth-fixed coordinate system, where x sat represents the x coordinate of the satellite's orbital position, y sat represents the y coordinate of the satellite's orbital position, and z sat represents the z coordinate of the satellite's orbital position.

[0068] Therefore, according to the user's requirements and the orbital altitude of the satellite, determine the distance d between two RFIs and the ground pixel resolution Res. When d ≤ Res, it is a close-range RFI source; when d > Res, it is a long-range RFI source.

[0069] S2: Perform flat-scene background adaptive RFI intensity determination

[0070] S2.1: For long-range RFI sources, adopt the independent point target intensity determination method, and iteratively suppress the influence of the background scene on the RFI source intensity estimation to obtain the accurate RFI source intensity

[0071] S2.1.1: Perform brightness temperature inversion on the original visibility measured by the synthetic aperture radiometer to obtain the original brightness temperature image T L1A (ξ,η) = T scence (ξ,η) + T RFI (ξ,η). Among them, T scence (ξ,η) represents the scene background brightness temperature, T RFI (ξ,η) represents the brightness temperature of the RFI source, and T L1A (ξ,η) represents the scene brightness temperature.

[0072] S2.1.2: In the original brightness temperature image (which is composed of the scene background brightness temperature and the RFI source brightness temperature), find the position (ξ RFI,m ,η RFI,m ) and intensity of the m-th brightness temperature peak If (selected as 350K according to the maximum radiation brightness temperature law of natural radiation), then there is an RFI source at the position (ξ RFI,m ,η RFI,m ), otherwise there is no RFI source in the scene.

[0073] S2.1.3: If there is an RFI source, calculate the visibility V of the RFI source at 1K response at (ξ RFI,m ,η RFI,m ) as V 1K (u,v,ξ RFI,m, η RFI,m ) and the 1K inversion brightness temperature of the RFI source

[0074] S2.1.4: According to (ξ RFI,m , η RFI,m ) calculate the peak brightness temperature of the scene at the position of the RFI source If the loop factor γ is taken as 0.25, then from and the intensity I of the RFI source is obtained RFI as:

[0075]

[0076] S2.1.5: Substitute the 1K inversion brightness temperature V of the RFI source 1K (u, v, ξ RFI,m , η RFI,m ) and the intensity I of the RFI source RFI into formula (1), and the scene visibility after suppressing the m-th RFI is formula (2):

[0077]

[0078] S2.1.6: Invert the visibility of the brightness temperature value, and repeat to find the RFI source and suppress the RFI source according to steps S2.1.2 - S2.1.3 until there are no point sources with a brightness temperature value greater than 350K in the brightness temperature image.

[0079] S2.1.7: Generate a list of determined positions of the RFI sources {(ξ RFI,1 , η RFI,1 ), …, (ξ RFI,q , η RFI,q )}

[0080] S2.1.8: Suppress each RFI source in the RFI list according to the following method.

[0081] 1) Calculate the average scene brightness temperature after suppressing the m-th RFI source. Taking (ξ RFI,m , η RFI,m ) as the center, calculate the average brightness temperature of a circular area with a radius r equal to 9 pixels. At this time, there are P pixels in the circular area:

[0082]

[0083] 2) Calculate the ratio ρ of the brightness temperature m after suppressing the RFI source to the average scene brightness temperature, where represents the inverse Fourier transform of the visibility function:

[0084]

[0085] 3) When ρ m <1, it indicates that the influence of the m-th RFI source on the scene has been completely suppressed. Then, the RFI intensity formula as shown in Equation (5) can be obtained, and the suppression of the (m + 1)-th RFI source can be carried out.

[0086]

[0087] 4) When ρ m >1, continue to iteratively suppress the m-th RFI source, that is:

[0088]

[0089] S2.2: For the step-by-step RFI intensity determination method for short-distance RFI sources, to solve the influence of adjacent high-intensity RFI sources on other RFI sources, in the logical order from strong to weak, estimate the intensity of RFI sources one by one, effectively weakening the potential influence of strong sources on the intensity of weak sources, thus ensuring the accuracy and reliability of the intensity determination result.

[0090] Brightness temperature image of the scene with RFI contamination can be expressed as:

[0091]

[0092] where:

[0093]

[0094] N T represents the total number of Y-shaped array antennas, and is a weighting function used to gradually reduce the visibility samples to reduce the influence of sidelobes. The brightness temperature of the scene affected by the point (ξ', η') Brightness temperature of the RFI source affected by the point (ξ', η') Equivalent array factor AF(ξ - ξ', η - η'), k antenna normalized voltage pattern l antenna normalized voltage pattern Weighting function w(u kl , v kl ) that gradually reduces the sampling of sidelobe visibility function.

[0095] The idea of step-by-step RFI intensity determination is to suppress the intensity contribution of RFI at the estimated position from the observation data, so as to extract the position and intensity of the signal source. Repeat the determination of the position and intensity of RFI in this way until repeatedly searching for the amplitude intensity of RFI to minimize Equation (11):

[0096]

[0097] In the step-by-step RFI strength determination method, in each iteration process, the position with the strongest brightness temperature peak is found in the brightness temperature image of the scene with RFI contamination as the position of the m-th RFI (ξ' RFI,m , η' RFI,m ) and intensity. Then, subtract from the brightness temperature image of the scene contaminated by RFI the influence factor AF(ξ - ξ' RFI , η - η' RFI ) centered at the point (ξ RFI , η RFI )·An·γ A n , the peak intensity An, the step factor γ, and the equivalent array factor AF(ξ - ξ' RFI , η - η' RFI ). After removing part of the RFI sources, the brightness temperature As shown in Equation (12):

[0098]

[0099] In the RFI processing of the step-by-step RFI strength determination method, by repeatedly finding the brightness temperature points higher than 350K in the image, iteratively suppressing the affected beams centered at the position of this point with the brightness temperature as the amplitude and multiplied by the step factor until the brightness temperature value of this point is lower than 350K. The accuracy of the RFI position estimation directly determines the performance of the step-by-step RFI strength determination method. However, if there are multiple RFI sources close to each other in the scene, or there are strong sources in the scene, due to the relatively large sidelobe response of the synthetic aperture radiometer to strong sources, the position estimation deviation of the RFI source is relatively large, thus reducing the positioning and strength determination accuracy of the RFI source. In addition, the selection of the step factor will also affect the entire iterative suppression effect.

[0100] The essence of the step-by-step RFI strength determination method is to improve the strength determination accuracy of the RFI source by improving the step factor and termination condition of the above formula. This method changes the step factor by changing the abnormal fluctuation of the RFI brightness temperature in the scene, thereby improving the number of loop iterations. At the same time, by using the standard deviation (STD) of the brightness temperature near the RFI position to replace the traditional fixed brightness temperature threshold, the termination suppression process is dynamically improved. Compared with the RFI intensity estimation of point targets, this method improves the efficiency of the entire suppression process through adaptive step factor iteration suppression, and thus can obtain better suppression effects in non-flat scene backgrounds (such as the land-sea boundary).

[0101] S3: Implement adaptive RFI strength determination for non-flat scene backgrounds

[0102] S3.1: For long-distance RFI sources in non-flat scenes, adopt the RFI strength determination method near the coastline to solve the RFI strength determination affected by the Gibbs oscillation at the land-sea boundary of RFI and the RFI source.

[0103] Bright temperature of the actual scene It can be composed of the bright temperature of the scene background and the bright temperature of the RFI source as shown in Formula (13):

[0104]

[0105] where, T V1K (ξ n , η n ) represents the 1K impulse response bright temperature of the point source at the position (ξ n , η n ). For the two-dimensional synthetic aperture radiometer, the 1K impulse response bright temperature T n (ξ n , η V1K )(ξ n , η n ) = F -1 [V 1K (u, v)].

[0106]

[0107] where, V scene (ξ, η) represents the visibility function at the scene background (ξ, η), and the visibility of the nth RFI source V L1A (ξ, η) represents the scene visibility function at (ξ, η), which can be obtained by calculating according to Formula (13);

[0108]

[0109] where, and respectively represent the maximum bright temperature value and the minimum bright temperature value of the RFI, and represent the maximum value and the minimum value of the average of 13×13 pixel points around the RFI (ξ n , η n ) in the scene as the center.

[0110] The average bright temperature of the scene after suppressing the RFI source can be expressed by the following formula

[0111]

[0112] Suppose the RFI source vector to be estimated is Formula (18), and the set of all RFI point sources is denoted as

[0113] {Μ(ξ1, η1), Μ(ξ2, η2),..., Μ(ξ n , ηn )}。

[0114]

[0115] The standard deviations of land and ocean are shown in Formulas (19) and (20).

[0116]

[0117] Therefore, the cost function J(Θ) represented by the overall fluctuation can be determined by the lower standard deviations of land and ocean BTs.

[0118]

[0119] where N land and N sea represent the numbers of RFI interferences on land and ocean. The energy of

[0120] The ultimate goal of the RFI intensity determination method near the coastline is to find the minimum value of Formula (21), i.e., Formula (22):

[0121]

[0122] S3.2: Decoherent RFI intensity determination method. By comparing the differences between the original brightness temperature and the retrieved brightness temperature, a system of equations with the RFI source intensity as the unknown is constructed, and the intensity of the coastal short-distance RFI source is determined by solving the system of equations.

[0123] Separate the influence of the scene Gibbs oscillation and the RFI source from the retrieved brightness temperature image, so as to obtain the true RFI energy intensity. With the help of the brightness temperature model, the relationship between the RFI source intensity and the Gibbs error can be obtained as shown in Formula (23).

[0124]

[0125] The right side of the formula consists of the Gibbs error of the earth scene and the Gibbs oscillation term with the RFI source intensity jointly, represents the scene retrieved brightness temperature, represents the true scene brightness temperature, AF w (ξ, η, ξ n , η n ) is the synthetic aperture radiometer array factor, and δ(ξ - ξ n , η - η n ) is the impulse response function located at (ξ n , η n ).

[0126] If the difference between the brightness temperature inverted from the Earth scene background and the true brightness temperature of the Earth scene background is approximately canceled out, Equation (23) can be simplified as:

[0127]

[0128] The scene inversion mode is obtained through the convolution operation of the scene simulated brightness temperature model and the array factor AF w (ξ,η,ξ′,η′), as shown in Equation (25):

[0129]

[0130] where, is the brightness temperature obtained based on model inversion. Thus, the scene Gibbs effect model is obtained as shown in Equation (26):

[0131]

[0132] When the RFI is located in the coastal area, subtract

[0133]

[0134] from both sides of Equation (24). The above-mentioned coastal area Gibbs term After elimination, its expression can be simplified as:

[0135]

[0136] Thus, an ideal brightness temperature image without the influence of the coastal area Gibbs effect is obtained.

[0137] Since the true scene brightness temperature is unknown, convolve AF w (ξ,η,ξ′,η′) on both sides. Let:

[0138] AF w (ξ,η,ξ′,η′)*AF w (ξ,η,ξ′,η′) =AF 2w (ξ,η,ξ′,η′) (29)

[0139] Convolve the array factor AF w (ξ,η,ξ′,η′) on both sides of Equation (24) to get:

[0140]

[0141] Substitute Equation (30) into Equation (24) to obtain Equation (31):

[0142]

[0143] Similar to the above situation, when the field of view is in the coastal area, convolving both sides of formula (24) with the array factor gives:

[0144]

[0145] The above formulas (31) and (32) respectively represent the equations for inland (sea) and coastal areas, and the mathematical expression between the intensities of multiple RFI sources and the model-inverted brightness temperature is achieved through the array factor.

[0146] Convolve both sides of the above formulas (31) and (32) with the array factor AF w (ξ, η, ξ′, η′), a window function with W≠1 must be multiplied. Otherwise, formula (29) is expressed as (33):

[0147] AF w (ξ, η, ξ′, η′)*AF w (ξ, η, ξ′, η′) =AF w (ξ, η, ξ′, η′) (33)

[0148] The brightness temperature convolution of the above formula with the array factor is equivalent to sampling at the spatial frequency domain W pq If the window function processing is not carried out, it will inevitably lead to the same sampling results obtained from two convolutions and one convolution. At this time, there will be a situation where both sides of formula (30) are always equal to 0.

[0149]

[0150] Therefore, in order to avoid the situations of formulas (33) and (34), the present invention uses the windowed array factor to implement the equation operation.

[0151] To solve for the intensity values of N RFI sources, N homogeneous equations with unknowns are required. Substituting different positions (ξ, η) into the equation can generate different equations, thus generating an N - yuan linear homogeneous equation system. However, usually, to reduce the calculation error, (ξ, η) = (ξ n , η n ) is selected and substituted into formula (31) or formula (32) in different cases. Usually in the case of the approximate solution is

[0152] The approximate equation system for the convolution operation in inland (sea) is:

[0153]

[0154] The approximate equations for convolution operations in coastal area scenarios are as follows:

[0155]

[0156] When the approximate intensity of the RFI source is obtained by solving After that, substituting the approximate solution of the RFI source into formula (37), the true inversion brightness temperature value of the RFI source in the scenario can be calculated, as shown in formula (38).

[0157]

[0158] The present invention will be further described below through specific embodiments:

[0159] (1) A simulation one-dimensional scenario background without RFI sources is used as the ideal scenario background of the one-dimensional synthetic aperture radiometer. RFI sources with different characteristics are added to the ideal scenario of the one-dimensional synthetic aperture radiometer. The RFI sources are divided into weak RFI sources (less than 350K), medium RFI sources (350K - 1000K), strong RFI sources (1000K - 5000K), and extremely strong RFI sources (greater than 5000K) to construct a simulated RFI scenario. Then, five different methods, namely the independent point target calibration method, the step-by-step RFI calibration method, the RFI calibration method near the coastline, the decoherence RFI calibration method, and the adaptive RFI calibration method (the calibration method of the present invention), are used to conduct comparative analysis of the calibration of this scenario respectively. The simulation analysis verification process is as Figure 3 shown.

[0160] (2) The one-dimensional synthetic aperture radiometer selects a pure ocean scenario and a land-sea scenario, and simulates and analyzes four cases: an independent weak source in the ocean scenario, a nearby extremely strong source in the land-sea scenario, an independent weak source in the land-sea scenario, and a nearby weak source in the land-sea scenario. Finally, a complex case with two nearby RFI sources, a strong and a weak RFI source, and an RFI source near the coastline is constructed, and performance analysis and verification are carried out for a total of five cases.

[0161] (3) RFI sources with different characteristics are added to the ideal ocean scenario and the ideal land-sea scenario background respectively to construct a simulated RFI scenario.

[0162] In the complex case of two nearby RFI sources, a strong and a weak RFI source, and an RFI source near the coastline in the land-sea scenario, the RFI source positions are: RFI_position = [-0.54 -0.36 0 0.03 0.23 0.28]; the RFI source intensities are: RFI_intensity = [500 3000 1500 1500 600 2300].

[0163] (4) Analyze and verify five RFI calibration methods, namely the independent point target calibration method (Method 1, abbreviated as D1), the step-by-step RFI calibration method (Method 2, abbreviated as D2), the RFI calibration method near the coastline (Method 3, abbreviated as D3), the decorrelation RFI calibration method (Method 4, abbreviated as D4), and the adaptive RFI calibration method proposed in this invention (the method of this invention, abbreviated as D5). The mean square error of the one-dimensional synthetic aperture radiometer for calibrating the RFI source is shown in Table 1 as follows:

[0164] Table 1 Calibration errors of different calibration methods of the one-dimensional synthetic aperture radiometer in different simple scenarios

[0165]

[0166] The mean square error of the RFI source of the one-dimensional synthetic aperture radiometer in a complex scenario is shown in Table 2 as follows:

[0167] Table 2 RFI calibration errors of different calibration methods of the one-dimensional synthetic aperture radiometer in a complex scenario

[0168]

[0169]

[0170] According to Table 1, in the application of the one-dimensional synthetic aperture radiometer, for a weak RFI source at a long distance in a flat pure ocean environment (Scenario 1, as shown in Figure 4 ), the calibration error of the independent point target calibration algorithm is the highest, reaching 1.5 K. Subsequently, the RFI calibration algorithm near the coastline has an error of 1.4 K. However, the adaptive RFI calibration algorithm is on a par with the distributed RFI calibration algorithm and the decorrelation RFI calibration algorithm, all achieving a relatively low error level of 1.3 K, demonstrating the superior performance of the adaptive RFI calibration algorithm in this environment; in the scenario of a strong RFI source at a short distance (Scenario 2, as shown in Figure 5 ), the decorrelation RFI calibration algorithm and the adaptive RFI calibration algorithm jointly reach the minimum error value, both being 2.0 K, showing their high stability and accuracy in a strong interference environment. For the scenario of a weak RFI source at a long distance (Scenario 3, as shown in Figure 6 ), the RFI calibration algorithm near the coastline and the adaptive RFI calibration algorithm are tied for the lead again, with errors of 1.3 K, while the decorrelation RFI calibration algorithm follows closely with an error of 1.6 K. And in the scenario of a weak RFI source at a short distance (Scenario 4, as shown in Figure 7As shown in [Figure 0], the decoherence RFI intensity determination algorithm and the adaptive RFI intensity determination algorithm continue to maintain the advantage of low error, with the error being 1.8K for both. However, the error of the two-point target intensity determination algorithm increases significantly to 16.8K, indicating poor performance. In summary, whether in a flat pure ocean environment or a non-flat scene at the land-sea boundary, in the face of 4 different RFI source distribution situations, the adaptive RFI intensity determination algorithm can achieve the minimum intensity determination error in the application of a one-dimensional synthetic aperture radiometer, fully demonstrating its universality in the application of a one-dimensional synthetic aperture radiometer.

[0171] According to Table 2, in the application of a one-dimensional synthetic aperture radiometer, in the face of a complex scene (Scene 5, as Figure 8 shown), when strong and weak sources appear simultaneously, for the weak source, due to being affected by the strong source, the intensity determination error of the weak source is larger than that of the strong source. However, the algorithm with the smallest intensity determination error is the adaptive RFI intensity determination algorithm, with the error of the weak source being 3.3K and that of the strong source being 2.8K. In the scene of a close-range RFI source at the land-sea boundary, in the worst case, the intensity determination error of the independent point target RFI intensity determination algorithm is more than 4 times that of the adaptive RFI intensity determination error, and the minimum error of the adaptive RFI intensity determination algorithm is 2.8K. In the scene where strong and weak sources coexist at a close range on land, the basic rule is that the intensity determination algorithm error is the largest near the coastline, and the best one is still the adaptive RFI intensity determination algorithm, with the minimum error being 1.8K. In summary, for a one-dimensional synthetic aperture radiometer, in a complex scene, the adaptive RFI intensity determination algorithm can always show the smallest intensity determination error, fully demonstrating its excellent performance and stability in the application of a one-dimensional synthetic aperture radiometer.

[0172] The above-described embodiments are only relatively preferred specific embodiments of the present invention. The ordinary changes and substitutions made by those skilled in the art within the scope of the technical solution of the present invention should be included in the protection scope of the present invention.

Claims

1. An adaptive radio frequency interference intensity determination method for a microwave radiometer, characterized in that, Including: Based on the brightness temperature image of the snapshot contaminated by RFI, perform adaptive scene analysis and judgment: classify the background areas where RFI sources are located on the brightness temperature image into flat scene backgrounds and non-flat scene backgrounds; pair up all RFI sources, and determine the distance between any pair of RFI sources and the ground pixel resolution according to user requirements and the orbital altitude of the satellite. If the distance does not exceed the ground pixel resolution, it is identified as a close-range RFI source, otherwise it is a long-range RFI source; For RFI sources in flat scene backgrounds, the methods for obtaining the RFI source intensity include: Process each pair of RFI sources as follows: if it is a close-range RFI source, use the step-by-step RFI intensity determination algorithm to obtain the RFI source intensity; if it is a long-range RFI source, use the independent point target intensity determination algorithm to obtain the RFI source intensity; For RFI sources in non-flat scene backgrounds, the methods for obtaining the RFI source intensity include: Process each pair of RFI sources as follows: if it is a long-range RFI source, use the RFI intensity determination algorithm near the coastline to obtain the RFI source intensity; if it is a close-range RFI source, use the de-coherent RFI intensity determination algorithm to obtain the RFI source intensity; If multiple intensity data are obtained for the same RFI source, the accurate intensity is obtained by taking the average.

2. The adaptive radio frequency interference intensity determination method for a microwave radiometer according to claim 1, wherein Classify the background areas where RFI sources are located on the brightness temperature image into flat scene backgrounds and non-flat scene backgrounds. The specific method is: According to the position (ξ i , η i ) of the i-th RFI source in the brightness temperature image, project (ξ i , η i ) onto the Earth coordinate system to obtain the longitude and latitude coordinates of the RFI source as O(lon i , lat i ); with O(lon i , lat i ) as the center, the area delineated by a circle with a set distance r as the radius is the buffer zone. If the number of intersection points between the buffer zone and the coastline γ curve is greater than or equal to 2, it is determined to be a coastal area, which belongs to the RFI source on the non-flat scene background; If the number of intersection points between the buffer zone and the γ curve of the coastline is less than 2, it is determined to be an inland sea or inland area, which belongs to the RFI source on the flat scene background.

3. The adaptive radio frequency interference intensity determination method for a microwave radiometer according to claim 1, wherein Determine the distance between two RFI sources according to user requirements and the orbital altitude of the satellite. The determination method is: Obtain the initial positioning coordinates of the RFI source in the antenna field-of-view coordinate system according to the brightness temperature threshold method; combine the orbital altitude and attitude information of the satellite to convert the initial positioning coordinates in the synthetic aperture radiation coordinate system to the coordinates in the Earth's longitude and latitude coordinate system; Calculate the distance d between the two RFI sources: d = R * arccos[cos(lon RFIi ) * cos(lon RFIj ) * cos(lat RFIi - lat RFIj ) + sin(lat RFIi ) * sin(lat RFIj )] where R is the radius of the Earth, (lon RFIi , lat RFIi ) are the longitude and latitude of the i-th RFI source, and (lon RFIj , lat RFIj ) are the longitude and latitude of the j-th RFI source.

4. A method for adaptively determining the intensity of radio frequency interference of a microwave radiometer according to claim 1, characterized in that, Determine 2 ground pixel resolutions Res according to user requirements and the orbital altitude of the satellite. The Res determination method is: Among them, D xmax represents the maximum baseline of the synthetic aperture radiometer antenna array in the X direction in the geocentric coordinate system, D ymax represents the maximum baseline distance of the synthetic aperture radiometer antenna array in the Y direction in the geocentric coordinate system, (x sat , y sat , z sat ) represents the orbital position information of the satellite in the geocentric coordinate system.

5. A method for adaptively determining the intensity of radio frequency interference in a microwave radiometer according to claim 1, characterized in that, For long-range RFI sources in flat scene backgrounds, use the independent point target intensity determination algorithm to obtain the RFI source intensity. The specific method is: 1) Perform brightness temperature inversion on the original visibility measured by the synthetic aperture radiometer to obtain a brightness temperature image, and find the positions and intensities of the brightness temperature peaks in the brightness temperature image; 2) Judge each brightness temperature peak in turn: if the brightness temperature peak intensity does not exceed 350K, there is no RFI source in the scene, and step 4) is executed, otherwise there is an RFI source at the position; 3) For each existing RFI source, perform the following processing: Calculate the 1K response visibility and 1K inversion brightness temperature of each RFI source to obtain the intensity of each RFI source; Calculate the scene visibility after suppressing all RFI sources according to the intensity of each RFI source, and return to step 1); 4) Generate a list of determined positions of RFI sources {(ξ RFI,1 , η RFI,1 ), …, (ξ RFI,q , η RFI,q ), and determine the strength of each RFI source in the RFI list one by one: Calculate the average scene brightness temperature after suppressing the current RFI source, and then calculate the ratio of the brightness temperature after suppressing the RFI source to the average scene brightness temperature; when the ratio is less than 1, it indicates that the influence of the current RFI source on the scene has been completely suppressed, and the intensity of the current RFI can be obtained; otherwise, continue to iteratively suppress the current RFI source until the ratio is less than 1, and then obtain the intensity of the current RFI.

6. A method for adaptively determining the intensity of radio frequency interference in a microwave radiometer according to claim 1, characterized in that, For a short-distance RFI source in a flat scene background, a step-by-step RFI intensity determination algorithm is adopted to obtain the RFI source intensity. The specific method is as follows: Suppress the intensity contribution of the RFI at the estimated position from the observation data, so as to extract the position and intensity of the signal source. Repeat the position determination and intensity determination of the RFI in this way until the amplitude intensity of the RFI is repeatedly searched 7. A method for adaptively determining the intensity of radio frequency interference in a microwave radiometer according to claim 1, characterized in that, For a distant RFI source in a non-flat scene background, use the RFI intensity determination algorithm near the coastline to obtain the RFI intensity determination affected by both the Gibbs oscillation at the land-sea boundary and the RFI source. The specific method is as follows: Calculate the average scene brightness temperature after suppressing the RFI source based on the actual scene brightness temperature composed of the scene background brightness temperature and the RFI source brightness temperature. Based on the average scene brightness temperature after suppressing the RFI source, minimize the following formula to obtain the final goal of the RFI intensity determination method near the coastline and achieve intensity determination: Where N land , N sea represent the number of RFI interferences on land and sea; σ land , σ sea represent the standard deviations of land and sea, which are obtained from the average brightness temperature of the scene after suppressing RFI sources.

8. A method for adaptively determining the intensity of radio frequency interference in a microwave radiometer according to claim 1, characterized in that, For a near-distance RFI source in a non-flat scene background, use the decoherence RFI intensity determination algorithm. By comparing the difference between the original brightness temperature and the inverted brightness temperature, a system of equations with the RFI source intensity as the unknown is constructed, and the intensity determination of the near-distance RFI source is achieved by solving the system of equations.

9. A computer-readable storage medium storing a computer program, characterized in that, When the described computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 8.

10. An electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the described computer program, it implements the steps of the method according to any one of claims 1 to 8.

Images